Debugging with the scientific method
Debugging is hard - this is of course known and rather obvious. Still, it is often approached on a mostly hunch-based basis. But we can do better - ideally, we want a system in which we can find our problem every time, and in a predictably finite manner. As it turns out that finding bugs in a large, unpredictable codebase tends to have a lot in common with looking for answers in a scientific research, we can borrow from the ideas and methods of that community to achieve just that. We’ll also talk about making your code debuggable in the first place, and possibly about rubber ducks.